THE LIMITATION FOR THE PHENOMENOLOGICAL CRYSTALLOGRAPHIC THEORY OF MARTENSITE TRANSFORMATIONS

Nanju Gu, Huifen Peng, Xiaoyan Song, Fuxing Yin and Yuguo Wang

Department of Material Science and Engineering, Hebei University of Technology Tianjin 300130, P.R.China

Keywords: martensite transformation, phenomenological theory, crystallography

The phenomenological crystallographic theory for martensitic transformations which is based on the principle of the invariant plane, has successfully predicted the habit plane, the orientation relationship, the shape deformation and the substructure of martensites in Au-Cd, In-Ti, Cu-Zn, Cu-Zn-Al and Cu-Al-Ni alloys as well as the {3 10 15}f and {259}f martensites in Fe-base alloy. However, as sown by Christian[1]: it is difficult to explain the formation of the {225}f plate martensites and the lath martensites in steels.

As pointed out by the authors[2,3,4], the formation of an invariant habit pane is connected with self-accommodation between different martensitic variants; and it is impossible to have self-accommodation in the lath martensite with a single variant, so the strain energy is mainly reduced by the plastic accommodation which would result in more rotation (about 130 ) of the interface (i.e. habit plane). That is why it is difficult to explain the {557}f habit plane of lath martensites by means of phenomenological crystallographic theory.

Two things account for the occurrence for the above limitations: First, the experimental evidence of the invariant habit plane both for WLR and BM theory, all was obtained from the twinning martensites in a Fe-22Ni-0.8C alloy[5], for which there is good self-accommodation between the different martensitic variants; while for the dislocation martensites no report has been presented relative to the evidence for the invariant plane.

Second, the original WLR theory[6] has pointed out incisively that the interface between the two phases be one of zero average distortion over a twin-related structure of martensite, since the Bain contraction axes for twin and matrix regions were original 900 from each other, so any vector in this interface extended by twin region and will be shortened by region of matrix, thus the interface is a macro-invariant plane. Regrettably, in their same paper they again proposed that the two Bain distortions with different contracted axes in twin and matrix regions, may be replaced by one in which regions 1 and 2 subjected to the same Bain distortion coupled with slip in region 2¡±, thus resulted the limitation of the WLR theory.

[1] J.W. Christian: Proc. of ICOMAT-86, Nara, Japan (1986)8.
[2] Nanju Gu et al.: Metall. and Mater. Trans. 26A (1995)1979.
[3] Nanju Gu et al.: Acta Metall. Sinica, 9(4) 1996;298-306.
[4] Nanju Gu et al. ISIJ International, 37(1997) No. 9;914-920.
[5] A.B.Greninger and A.R.Troiano: Trans.AIME, 185 (1949) 591
[6] M.S.Wechsler et al.: Trans .AIME, 11 (1953) 1503.